检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Guo, Yihan Liu, Tingwen Sheng, Jiawei Ma, Duohe Sun, Ming Tian, Ling Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798350368741

Document retrieval aims to find documents related to the query from all candidate documents. Existing studies develop the Generative Retrieval approach, which assigns a unique DocID to each document, and then measures document-query relevance based on the probability of generating the expected DocID for the given query. However, the generated DocID can have expressive limitation of different semantic topics, leading to semantic gaps in document retrieval. Besides, existing GR models usually suffer from the catastrophic forgetting when memorizing new documents incrementally, which makes real-world application impractical. To overcome these issues, we propose Translational Generative Retrieval, which translates each document into potential queries reflecting different topics, specifically by a sequence of token probability distributions. To better model the semantics of potential queries and effectively decode the target query from the distributions, we propose DirEcted Acyclic Graph Retrieval (DEAR) model, which reforms the distributions from a non-autoregressive generative model into a Directed Acyclic Graph. Experimental results demonstrate that DEAR outperforms existing retrieval models, setting a new state-of-the-art in generative retrieval. © 2025 IEEE.

关键词： Directed Acyclic Graph Generative Retrieval Non-autoregressive Generation Query Generation

来源：评论

学校读者我要写书评

暂无评论

Cleaning of two mirrors in the first mirror unit using radiofrequency capacitively coupled plasma

引用

Plasma science and technology 2024年第9期26卷 165-171页

作者： Chenxue WANG Rong YAN Yuming LIU Su XU Lei MU Wei ZHENG Rui DING Junling CHEN Institute of Plasma Physics Chinese Academy of SciencesHefei 230031People’s Republic of China University of Science and Technology of China Hefei 230026People’s Republic of China School of Computer Engineering Anhui Sanlian UniversityHefei 230601People’s Republic of China

First mirror(FM)cleaning,using radio frequency(RF)plasma,has been proposed to recover FM reflectivity in nuclear fusion reactors such as the International Thermonuclear Experimental Reactor(ITER).To investigate the influence of simultaneous cleaning of two mirrors on mirror cleaning efficiency and uniformity,experiments involving single-mirror cleaning and dual-mirror cleaning were conducted using RF capacitively coupled plasma in the *** the test and simultaneous cleaning of two mirrors,the FM and second mirror(SM),both measuring 110 mm×80 mm,were placed inside the first mirror unit(FMU).They were composed of 16 mirror samples,each with a dimension of 27.5 mm×20 *** mirror samples consist of a titanium-zirconium-molybdenum alloy substrate,a 500 nm Mo intermediate layer and a 30 nm Al_(2)O_(3) surface coating as a proxy for Be *** cleaning of a single first mirror(SFM)and the simultaneous cleaning of the FM and SM(DFM and DSM)lasted for 9 h using Ar plasma at a pressure of 1 *** total reflectivity of mirror samples on the DSM did not fully recover and varied with location,with a self-bias of−140 *** a self-bias of−300 V,the total reflectivity of mirror samples on the SFM and DFM was fully *** energy dispersive spectrometer results demonstrated that the Al_(2)O_(3) coating had been completely removed from these mirror ***,the mass loss of each mirror sample on the SFM and DFM before and after cleaning varied depending on its location,with higher mass loss observed for mirror samples located in the corners and lower loss for those in the *** with SM cleaning,the simultaneous cleaning of two mirrors reduced the difference between the highest and lowest mass ***,this mass loss for the mirror samples of the DFM facing the DSM was *** indicated that mirror samples cleaned face to face in the FMU simultaneously could influence each other,highlighting the need for special attention in

关键词： first mirror unit dual-mirror cleaning reflectivity sputtering rate

来源：评论

学校读者我要写书评

暂无评论

MIWE: Multimodal Indexing of Web Entities Incorporating Semantic Artificial Intelligence

MIWE: Multimodal Indexing of Web Entities Incorporating Sema...

引用

International Conference on Data science, Computation and Security, IDSCS 2023

作者： Aravind Krishnan, A. Deepak, Gerard Department of Computer Science and Engineering Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India

ISBN: (纸本)9789819709748

Web indexing is of critical importance to Web 3.0 owing to the reason that it is not just expanding but growing exponentially. This paper proposes a multimodal indexing mechanism that indexes web images, videos, pages, and documents. The strategy encompasses paradigms like TF-IDF, ontology generation, and label extraction from different datasets. The datasets are subjected to the LSTM for classification for the web image and web video datasets and an LR classifier is used to classify the web document and web page dataset. Subsequently, the formalization of knowledge graphs and enhancement using Google’s knowledge graph API, context frees, microformat chain derivations, the encompassment of Wikidata and NELL knowledge stores, and web crawler agents increases the quantity of auxiliary knowledge, and the convergence to optimality. Semantic relevance is calculated using the NPMI measure. The MIWE shows improved performance when examined against other similar models. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Web crawler

来源：评论

学校读者我要写书评

暂无评论

Automating the Optimal Temperature to Increase the Yield in Sericulture by Using Internet of Things 5

Automating the Optimal Temperature to Increase the Yield in ...

引用

5th International Conference on Circuits, Control, Communication and Computing, I4C 2024

作者： Hemanth Chandra, N. Harsha, R. Prithvi Veera, B. Rohan, M.N. Thriveni, J. Venugopal, K.R. University Visvesvaraya College of Engineering Bangalore University Department of Computer Science and Engineering Bengaluru India Global Academy of Technology Department of Information Science and Engineering Bengaluru India Former Vice-Chancellor Bangalore University Bengaluru India

ISBN: (数字)9798331528539

ISBN: (纸本)9798331528539

Sericulture/, the art of cultivating silk through the rearing of silkworms, earns its title as the Queen of Textiles owing to its unmatched softness, elegance, durability, and tensile strength. India, the second-largest producer of silk, favors this luxurious fabric for its exceptional water absorbency, heat resistance, and inherent luster. The intricacies of insect adaptation, particularly evident in mulberry silkworms, underscore the significance of maintaining optimal temperature and humidity levels, factors pivotal for their growth and survival. However, conventional methods of temperature regulation, such as employing wetted gunny bags, are insufficient for the precise monitoring demanded by sericulture. In this paper, a model has been proposed to revolutionize sericulture management by leveraging IoT and cloud technologies to ensure real-time monitoring of temperature and humidity within silkworm sheds. By offering insights into the interplay of humidity, airflow, and light on silkworm development, this system aims to enhance cocoon crop yields while minimizing the need for manual oversight. This holistic approach guarantees the production of superior-quality silk with consistent characteristics, heralding a new era of efficiency and excellence in the silk industry. © 2024 IEEE.

关键词： Silk

来源：评论

学校读者我要写书评

暂无评论

Support Vector Machines Based Predictive Seizure Care using IoT-Wearable EEG Devices for Proactive Intervention in Epilepsy 2

Support Vector Machines Based Predictive Seizure Care using ...

引用

2nd International Conference on computer, Communication and Control, IC4 2024

作者： Srinivas, Porandla Arulprakash, M. Vadivel, M. Anusha, N. Rajasekar, G. Srinivasan, C. Malla Reddy Institute of Engineering and Technology Department of Information Technology Telangana Hyderabad India Srm Institute of Science and Technology Department of Computing Technologies Kattankulathur Chenagalpattu Tamil Nadu Chennai India Vidya Jyothi Institute of Technology Department of Electronics and Communication Engineering Telangana Hyderabad India Vellore Institute of Technology School School of Computer Science and Engineering Department of IoT Tamil Nadu Vellore India Vinayaka Mission's Kirupananda Variyar Engineering College Department of Computer Science and Engineering Tamil Nadu Salem India Saveetha Institute of Medical and Technical Sciences Saveetha University Saveetha School of Engineering Department of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350387933

Epilepsy, a neurological illness that causes repeated seizures, can interfere with everyday life and needs prompt treatment. Internet of Things (IoT) wearable Electroencephalogram (EEG) devices and Support Vector Machines (SVM) for predictive analytics are used in this study to suggest a unique strategy for proactive seizure treatment. Wearable EEG devices feed real-time brain activity data. It uses SVM, a strong machine learning method, to build a prediction model using historical EEG data to identify and predict seizures. The prediction software analyses EEG data in real-time to detect pre-seizure patterns and initiate preventive treatments. The seizure prediction method uses SVM's capacity to handle high-dimensional data and catch complicated patterns to improve accuracy and reliability. Healthcare practitioners and caregivers may get timely warnings and react efficiently thanks to the IoT infrastructure's seamless connectivity between wearable devices and a centralized monitoring system. It discusses the ethical and privacy issues of installing such a system, stressing user permission and data protection. Pilot investigations show promising prediction accuracy and reaction time. SVM with IoT-wearable EEG sensors for predictive seizure care offers a forward-looking technique for enhancing epilepsy patients' quality of life by enabling individualized and proactive treatments. © 2024 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Zero-Shot Cross-Domain Slot Filling with Retrieval Augmented In-Context Learning

Zero-Shot Cross-Domain Slot Filling with Retrieval Augmented...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Song, Mengxiao Liu, Tingwen Li, Quangang Ma, Duohe Sun, Ming Tian, Ling Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798350368741

Zero-shot cross-domain slot filling is becoming increasingly important due to its ability to generalize to new domains without the need for annotating domain-specific data, which aligns well with the requirements of industrial deployments. Recent advanced works deal with this task through question answering framework and make remarkable progress. However, they always rely on human efforts to manually construct question templates or prompts for all slot types, which is not only labor consuming, but also experience context inconsistency issue between the manual example and the specific test instance. To alleviate this problem, we introduce a retriever designed to extract reference samples from the training sets, serving as demonstrations to guide the model in generating the target slot entity through in-context learning. Building upon this retriever, we propose a retrieval-augmented generative framework that automatically constructs and tailors prompts to each specific test instance, eliminating the need for manual efforts. Experiment results verify that our approach attains the state-of-the-art. © 2025 IEEE.

关键词： slot filling spoken language understanding zero-shot cross-domain

来源：评论

学校读者我要写书评

暂无评论

MLPDR: High Performance ML Algorithms for the Prediction of Diabetes Retinopathy

MLPDR: High Performance ML Algorithms for the Prediction of ...

引用

2023 IEEE International Conference on Network, Multimedia and Information technology, NMITCON 2023

作者： Madhuri G, Sindhu Chokkanathan, K. Mahesh, T.R. Musthafa, M.Mohamed Vanitha, K. Vivek, V. Manipal Institute of Technology Department of Information Technology Bengaluru India Manipal Academy of Higher Education Manipal India Madanapalle Institute of Technology & Science Department of Ai Madanapalle India Faculty of Engineering and Technology Department of Computer Science and Engineering Bangalore India Tamil Nadu Erode India Karpagam Academy of Higher Education Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798350300826

In today's world, millions of imaging analyses are performed every week, and medical imaging plays a significant role in that. It is a process of creating visual representation based on the functioning of human organs or tissues to identify abnormalities or study diseases. Diabetes has been regarded as one of the most prevalent diseases by many researchers. Blindness or other retinal alterations based on the eye's vision, such as difficulty reading or seeing distant things, may occur, which has an impact on diabetes in humans. A new computer aided diagnosis system based on image processing of retinal images can be analyzed in order to detect diabetic patients in advance. In this study, the eye retina based on image processing techniques is analyzed and machine learning techniques have been used to extract features of eye and several computational intelligence Machine Learning techniques like Support Vector Machines (SVM) and Principal Component Analysis (PCA). These Computational Intelligence techniques have been used to detect the diabetic in advance. Several performance metrics have been considered such as Root Mean Square Error (RMSE), Peak Signal Noise Ratio (PSNR), Entropy (E), and Accuracy (A) in obtaining the better accuracy results for detection of diabetic retinopathy in early stages. The experimental results show that, SVM outperforms with an accuracy of 95%. © 2023 IEEE.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

A probabilistic public key encryption switching scheme for secure cloud storage

引用

International Journal of Information technology (Singapore) 2023年第2期15卷 675-690页

作者： Bhat, Radhakrishna Sunitha, N.R. Iyengar, S.S. Department of Computer Science and Engineering Manipal Institute of Technology Manipal Academy of Higher Education Karnataka Manipal 576104 India Department of Computer Science and Engineering Siddaganga Institute of Technology Karnataka Tumkur 572103 India School of Computing and Information Science Florida International University Miami FL United States

The high demand for user-centric applications such as secure cloud storage laid the foundation for the development of user-centric security schemes with multiple security features in recent years. But, the current state-of-art techniques primarily emphasized only one type of security feature i.e., either homomorphism or non-malleability. To fill this gap and provide a common platform for both homomorphic and non-malleable cloud applications, we have introduced a new public key-based probabilistic encryption switching method. The proposed scheme generates O(m+2 log N) bits of ciphertext where m∈ N and m< n, n∈ N is the plaintext size, N is the RSA composite. The superior performance of the proposed scheme has been tested in comparison to existing methods and is reported in this paper. © 2022, The Author(s).

关键词： Encryption Switching Homomorphic encryption Probabilistic encryption Public key cryptosystem Secure Cloud storage and retrieval

来源：评论

学校读者我要写书评

暂无评论

Robust Sensor Networks: Applying AI for Fault Identification in Challenging Operational Contexts 5

Robust Sensor Networks: Applying AI for Fault Identification...

引用

5th International Conference on IoT Based Control Networks and Intelligent Systems, ICICNIS 2024

作者： Manjunatha, A.S. Venkatramana Bhat, P. Ramakrishna, M. Prabhu, Matti Nidhi Dept. of Computer and Communication Engineering NMAM Institute of Technology Nitte Karnataka Mangaluru India Dept. of Computer Science and Engineering NMAM Institute of Technology Nitte Karnataka Mangaluru India Manipal Academy of Higher Education Dept. of Data Science and Computer Applications Manipal Institute of Technology Manipal576104 India

ISBN: (纸本)9798331518097

Wireless Sensor Networks (WSNs) are highly implemented in harsh and dynamic environments. In such configurations, faults and anomalies easily occur, causing unwanted disruptions in data accuracy and network performance. In the context of reliability and stability in WSN operations, effective fault detection techniques are vital. This research work discusses how several machine learning models may be applied, including Logistic Regression, Decision Tree, Random Forest, and K-Nearest Neighbors in order to detect fault in WSNs. The performance evaluation was performed based on major criteria, which are accuracy, precision, recall, and Fl score. This, therefore, makes these approaches to machine learning effective in identifying faults efficiently, hence offering a well-balanced solution that is scalable and robust for integrity and operational effectiveness of WSN s. © 2024 IEEE.

关键词： Random forests

来源：评论

学校读者我要写书评

暂无评论

MLP-Based Speech Emotion Recognition for Audio and Visual Features 1

引用

7th International Conference on Microelectronics and Telecommunication engineering, ICMETE 2023

作者： Kothai, G. Boora, Prabhas Bhanu Muzammil, S. Venkata Subhash, L. Naga Raju, B. KPR Institute of Engineering and Technology Tamil Nadu Uthupalayam Coimbatore India Department of Computer Science and Engineering Kalasalingam Academy of Research and Education Virudhunagar Krishnankoil India

ISBN: (数字)9789819995622

ISBN: (纸本)9789819995615

Due to its potential applications in domains involving psychology, social smart machines, and human–computer interaction, speech emotion recognition has become an important and growing topic field in the study. That article offers the service comparative analysis of strategies in artificial intelligence classifiers for speech emotion recognition, including MLP, decision tree, SVM, and random forest. Using visual analysis methods like wave plot and spectrogram analysis as well as audio feature extraction, we evaluate these classifiers. According to our observations, MLP achieves better performance than the other classifiers, obtaining an accuracy of 85% when identifying different emotions. Moreover, we illustrate how audio feature extraction and visual analysis help to improve emotion recognition efficiency. Our research has applications for creating speech emotion recognition algorithms that may be applied in real-life scenarios. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：